This study investigates the hypothesis that inflammatory cytokines, interleukin (IL)-1alpha IL-1beta, and tumor necrosis factor (TNF), influence cardiac function by affecting calcium homeostasis and that this effect is mediated by the beta-adrenergic-adenylate cyclase system. After 4 days in culture, neonatal rat ventricular myocytes were treated with cytokines (10 ng/ml) for short (2 h) or longer (18 h) times. Myocyte calcium, contractility, and adenylate cyclase were measured under each condition. Anticipated stepwise increases in adenylate cyclase and intracellular calcium were found in controls (non-cytokine-treated) with 10(-7) M isoproterenol, 10(-7) M isoproterenol + 0.1 mM guanosine triphosphate, and 10(-9) M forskolin. Cells in the presence of cytokine for 2 h show increased basal calcium levels but no changes in adenylate cyclase activities, and isoproterenol fails to elevate adenylate cyclase levels or affect contractile shortening. After long-term treatment with IL-1beta or TNF, but not IL-1alpha, the significantly elevated levels of basal systolic calcium remain, and isoproterenol increases adenylate cyclase activity, unlike after short exposure. Forskolin maximally activates adenylate cyclase following both short- and long-term incubation, but the stepwise increase in activity is blunted following prolonged exposure. Thus short-term cytokine treatment blocks the adrenergic receptor-mediated increases in adenosine 3',5'-cyclic monophosphate, dissociating adenylate cyclase activation from cytokine-mediated increases in cell calcium, whereas longer treatment apparently produces direct affects on adenylate cyclase. Time-dependent differences in contractile response were found with IL-1alpha at 2 h and TNF at 18 h, implying that myofibrillar responsiveness to increased cytoplasmic calcium is dependent on both cytokine species and exposure time.